U.S. patent number 10,135,988 [Application Number 15/892,166] was granted by the patent office on 2018-11-20 for techniques for case allocation.
This patent grant is currently assigned to Afiniti Europe Technologies Limited. The grantee listed for this patent is Afiniti Europe Technologies Limited. Invention is credited to David J. Delellis, Ittai Kan.
United States Patent |
10,135,988 |
Kan , et al. |
November 20, 2018 |
Techniques for case allocation
Abstract
Techniques for case allocation are disclosed. In one particular
embodiment, the techniques may be realized as a method for case
allocation comprising receiving, by at least one computer
processor, at least one case allocation allocated using a first
pairing strategy, and then reassigning, by the at least one
computer processor, the at least one case allocation using
behavioral pairing.
Inventors: |
Kan; Ittai (McLean, VA),
Delellis; David J. (Doylestown, PA) |
Applicant: |
Name |
City |
State |
Country |
Type |
Afiniti Europe Technologies Limited |
Cheshire |
N/A |
GB |
|
|
Assignee: |
Afiniti Europe Technologies
Limited (Cheshire, GB)
|
Family
ID: |
58057172 |
Appl.
No.: |
15/892,166 |
Filed: |
February 8, 2018 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20180167513 A1 |
Jun 14, 2018 |
|
Related U.S. Patent Documents
|
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
Issue Date |
|
|
15364699 |
Nov 30, 2016 |
9924041 |
|
|
|
62261780 |
Dec 1, 2015 |
|
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H04M
3/5233 (20130101); G06Q 10/06311 (20130101); H04M
2203/408 (20130101) |
Current International
Class: |
H04M
3/523 (20060101); G06Q 10/06 (20120101) |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
|
|
|
|
|
|
|
2008349500 |
|
May 2014 |
|
AU |
|
2009209317 |
|
May 2014 |
|
AU |
|
2009311534 |
|
Aug 2014 |
|
AU |
|
2015203175 |
|
Jul 2015 |
|
AU |
|
102301688 |
|
May 2014 |
|
CN |
|
102017591 |
|
Nov 2014 |
|
CN |
|
0 493 292 |
|
Jul 1992 |
|
EP |
|
0 949 793 |
|
Oct 1999 |
|
EP |
|
1 032 188 |
|
Aug 2000 |
|
EP |
|
1335572 |
|
Aug 2003 |
|
EP |
|
11-098252 |
|
Apr 1999 |
|
JP |
|
2000-069168 |
|
Mar 2000 |
|
JP |
|
2000-078291 |
|
Mar 2000 |
|
JP |
|
2000-078292 |
|
Mar 2000 |
|
JP |
|
2000-092213 |
|
Mar 2000 |
|
JP |
|
2000-236393 |
|
Aug 2000 |
|
JP |
|
2001-292236 |
|
Oct 2001 |
|
JP |
|
2001-518753 |
|
Oct 2001 |
|
JP |
|
2002-297900 |
|
Oct 2002 |
|
JP |
|
3366565 |
|
Jan 2003 |
|
JP |
|
2003-187061 |
|
Jul 2003 |
|
JP |
|
2004-056517 |
|
Feb 2004 |
|
JP |
|
2004-227228 |
|
Aug 2004 |
|
JP |
|
2006-345132 |
|
Dec 2006 |
|
JP |
|
2007-324708 |
|
Dec 2007 |
|
JP |
|
2011-511533 |
|
Apr 2011 |
|
JP |
|
2011-511536 |
|
Apr 2011 |
|
JP |
|
5421928 |
|
Feb 2014 |
|
JP |
|
5631326 |
|
Nov 2014 |
|
JP |
|
5649575 |
|
Jan 2015 |
|
JP |
|
2015-514371 |
|
May 2015 |
|
JP |
|
316118 |
|
Dec 2013 |
|
MX |
|
322251 |
|
Jul 2014 |
|
MX |
|
587100 |
|
Oct 2013 |
|
NZ |
|
587101 |
|
Oct 2013 |
|
NZ |
|
591486 |
|
Jan 2014 |
|
NZ |
|
592781 |
|
Mar 2014 |
|
NZ |
|
1-2010-501704 |
|
Feb 2014 |
|
PH |
|
1-2010-501705 |
|
Feb 2015 |
|
PH |
|
WO-1999/017517 |
|
Apr 1999 |
|
WO |
|
WO-2001/063894 |
|
Aug 2001 |
|
WO |
|
WO-2006/124113 |
|
Nov 2006 |
|
WO |
|
WO-2009/097018 |
|
Aug 2009 |
|
WO |
|
WO-2010/053701 |
|
May 2010 |
|
WO |
|
WO-2011/081514 |
|
Jul 2011 |
|
WO |
|
Other References
Anonymous. (2006) "Performance Based Routing in Profit Call
Centers," The Decision Makers' Direct, located at
www.decisioncraft.com, Issue Jun. 2002 (3 pages). cited by
applicant .
Canadian Office Action issued in Canadian Patent Application No.
2713526, dated Oct. 25, 2016, 7 pages. cited by applicant .
Cleveland, William S., "Robust Locally Weighted Regression and
Smoothing Scatterplots," Journal of the American Statistical
Association, vol. 74, No. 368, pp. 829-836 (Dec. 1979). cited by
applicant .
Cormen, T. H., et al., "Introduction to Algorithms," 3rd Edition,
Chapter 26 Maximum Flow, pp. 708-768 and Chapter 29 Linear
Programming, pp. 843-897 (2009). cited by applicant .
Extended European Search Report issued by the European Patent
Office for European Application No. 17154781.3 dated May 4, 2017 (7
pages). cited by applicant .
Extended European Search Report issued by the European Patent
Office for European Application No. 17171761.4 dated Aug. 30, 2017
(8 pages). cited by applicant .
Gans, N. et al. (2003), "Telephone Call Centers: Tutorial, Review
and Research Prospects," Manufacturing & Service Operations
Management, vol. 5, No. 2, pp. 79-141. cited by applicant .
International Preliminary Report on Patentability issued in
connection with PCT Application No. PCT/US2009/066254 dated Jun.
14, 2011 (6 pages). cited by applicant .
International Search Report and Written Opinion issued by the
European Patent Office as International Searching Authority for
International Application No. PCT/IB2016/001762 dated Feb. 20, 2017
(15 pages). cited by applicant .
International Search Report and Written Opinion issued by the
European Patent Office as International Searching Authority for
International Application No. PCT/IB2016/001776 dated Mar. 3, 2017
(16 pages). cited by applicant .
International Search Report and Written Opinion issued by the
European Patent Office as International Searching Authority for
International Application No. PCT/IB2016/001871 dated Mar. 17, 2017
(13 pages). cited by applicant .
International Search Report and Written Opinion issued by the
European Patent Office as International Searching Authority for
International Application No. PCT/IB2017/000570 dated Jun. 30, 2017
(13 pages). cited by applicant .
International Search Report issued in connection with International
Application No. PCT/US13/33268 dated May 31, 2013 (2 pages). cited
by applicant .
International Search Report issued in connection with PCT
Application No. PCT/US/2009/054352 dated Mar. 12, 2010, 5 pages.
cited by applicant .
International Search Report issued in connection with PCT
Application No. PCT/US2008/077042 dated Mar. 13, 2009 (3 pages).
cited by applicant .
International Search Report issued in connection with PCT
Application No. PCT/US2009/031611 dated Jun. 3, 2009 (5 pages).
cited by applicant .
International Search Report issued in connection with PCT
Application No. PCT/US2009/066254 dated Feb. 24, 2010 (4 pages).
cited by applicant .
International Search Report issued in connection with
PCT/US2009/061537 dated Jun. 7, 2010 (5 pages). cited by applicant
.
International Search Report issued in connection with
PCT/US2013/033261 dated Jun. 14, 2013 (3 pages). cited by applicant
.
International Search Report issued in connection with
PCT/US2013/33265 dated Jul. 9, 2013 (2 pages). cited by applicant
.
Japanese Office Action issued by the Japan Patent Office for
Application No. 2015-503396 dated Jun. 29, 2016 (7 pages). cited by
applicant .
Japanese Office Action issued by the Japanese Patent Office for
Japanese Application No. 2016-159338 dated Oct. 11, 2017 (12
pages). cited by applicant .
Japanese Office Action issued by the Japanese Patent Office for
Japanese Application No. 2016-189126 dated Oct. 19, 2017 (24
pages). cited by applicant .
Koole, G. (2004). "Performance Analysis and Optimization in
Customer Contact Centers," Proceedings of the Quantitative
Evaluation of Systems, First International Conference, Sep. 27-30,
2004 (4 pages). cited by applicant .
Koole, G. et al. (Mar. 6, 2006). "An Overview of Routing and
Staffing Algorithms in Multi-Skill Customer Contact Centers,"
Manuscript, 42 pages. cited by applicant .
Nocedal, J. and Wright, S. J., "Numerical Optimization," Chapter 16
Quadratic Programming, pp. 448-496 (2006). cited by applicant .
Ntzoufras, "Bayesian Modeling Using Winbugs". Wiley Interscience,
Chapter 5, Normal Regression Models, Oct. 18, 2007, pp. 155-220 (67
pages). cited by applicant .
Press, W. H. and Rybicki, G. B., "Fast Algorithm for Spectral
Analysis of Unevenly Sampled Data," The Astrophysical Journal, vol.
338, pp. 277-280 (Mar. 1, 1989). cited by applicant .
Riedmiller, M. et al. (1993). "A Direct Adaptive Method for Faster
Back Propagation Learning: The RPROP Algorithm," 1993 IEEE
International Conference on Neural Networks, San Francisco, CA,
Mar. 28-Apr. 1, 1993, 1:586-591. cited by applicant .
Stanley et al., "Improving call center operations using
performance-based routing strategies," Calif. Journal of Operations
Management, 6(1), 24-32, Feb. 2008; retrieved from
http://userwww.sfsu.edu/saltzman/Publist.html. cited by applicant
.
Subsequent Substantive Examination Report ssued in connection with
Philippines Application No. 1-2010-501705 dated Jul. 14, 2014 (1
page). cited by applicant .
Substantive Examination Report issued in connection with
Philippines Application No. 1/2011/500868 dated May 2, 2014 (1
page). cited by applicant .
Written Opinion of the International Searching Aurhority issued in
connection with PCT Application No. PCT/US2008/077042 dated Mar.
13, 2009, 6 pages. cited by applicant .
Written Opinion of the International Searching Authority issued in
connection with International Application No. PCT/US13/33268 dated
May 31, 2013, 7 pages. cited by applicant .
Written Opinion of the International Searching Authority issued in
connection with PCT Application No. PCT/US/2009/054352 dated Mar.
12, 2010, 5 pages. cited by applicant .
Written Opinion of the International Searching Authority issued in
connection with PCT Application No. PCT/US2009/031611 dated Jun. 3,
2009, 7 pages. cited by applicant .
Written Opinion of the International Searching Authority issued in
connection with PCT Application No. PCT/US2009/066254 dated Feb.
24, 2010, 5 pages. cited by applicant .
Written Opinion of the International Searching Authority issued in
connection with PCT/US2009/061537 dated Jun. 7, 2010, 10 pages.
cited by applicant .
Written Opinion of the International Searching Authority issued in
connection with PCT/US2013/033261 dated Jun. 14, 2013, 7 pages.
cited by applicant .
Written Opinion of the International Searching Authority issued in
connection with PCT/US2013/33265 dated Jul. 9, 2013, 7 pages. cited
by applicant.
|
Primary Examiner: Hong; Harry S
Attorney, Agent or Firm: Wilmer Cutler Pickering Hale and
Dorr LLP
Parent Case Text
CROSS-REFERENCE TO RELATED APPLICATIONS
This patent application is a continuation patent application of
U.S. patent application Ser. No. 15/364,699, filed Nov. 30, 2016,
which claims priority to U.S. Provisional Patent Application No.
62/261,780, filed Dec. 1, 2015, each of which is hereby
incorporated by reference in its entirety as if fully set forth
herein.
Claims
The invention claimed is:
1. A method for benchmarking pairing strategies in a service center
system comprising: receiving, by at least one computer processor
communicatively coupled to and configured to operate in the service
center system, a first plurality of results for a first plurality
of cases assigned to a plurality of agents using a first pairing
strategy; receiving, by the at least one computer processor, a
second plurality of results for a second plurality of cases
assigned to the plurality of agents using a second pairing strategy
different from the first pairing strategy; determining, by the at
least one computer processor, a difference in performance between
the first and second pluralities of results, wherein the difference
in performance provides an indication that assigning cases using
the second pairing strategy results in a performance gain for the
service center system attributable to the second pairing strategy,
wherein the difference in performance also provides an indication
that optimizing performance of the service center system is
realized using the second pairing strategy instead of the first
pairing strategy; and outputting, by the at least one computer
processor, the difference in performance between the first pairing
strategy and the second pairing strategy for benchmarking at least
the first pairing strategy and the second pairing strategy.
2. The method of claim 1, wherein the first pairing strategy is
case assignment by a management authority.
3. The method of claim 1, wherein the first pairing strategy is a
first-in, first out (FIFO) pairing strategy.
4. The method of claim 1, wherein the second pairing strategy is a
behavioral pairing strategy to optimize performance of the service
center system.
5. The method of claim 1, wherein each case of the second plurality
of cases comprises a confidence level below a threshold confidence
level.
6. The method of claim 1, wherein each case of the first plurality
of cases comprises a rationale from a management authority and a
confidence level above a threshold confidence level.
7. The method of claim 1, wherein each case of the second plurality
of cases is randomly selected, by the at least one computer
processor, for assignment.
8. A system for benchmarking pairing strategies in a service center
system comprising: at least one computer processor communicatively
coupled to and configured to operate in the service center system,
wherein the at least one computer processor is configured to:
receive a first plurality of results for a first plurality of cases
assigned to a plurality of agents using a first pairing strategy;
receive a second plurality of results for a second plurality of
cases assigned to the plurality of agents using a second pairing
strategy different from the first pairing strategy; determine a
difference in performance between the first and second pluralities
of results, wherein the difference in performance provides an
indication that assigning cases using the second pairing strategy
results in a performance gain for the service center system
attributable to the second pairing strategy, wherein the difference
in performance also provides an indication that optimizing
performance of the service center system is realized using the
second pairing strategy instead of the first pairing strategy; and
output the difference in performance between the first pairing
strategy and the second pairing strategy for benchmarking at least
the first pairing strategy and the second pairing strategy.
9. The system of claim 8, wherein the first pairing strategy is
case assignment by a management authority.
10. The system of claim 8, wherein the first pairing strategy is a
first-in, first out (FIFO) pairing strategy.
11. The system of claim 8, wherein the second pairing strategy is a
behavioral pairing strategy to optimize performance of the service
center system.
12. The system of claim 8, wherein each case of the second
plurality of cases comprises a confidence level below a threshold
confidence level.
13. The system of claim 8, wherein each case of the first plurality
of cases comprises a rationale from a management authority and a
confidence level above a threshold confidence level.
14. The system of claim 8, wherein each case of the second
plurality of cases is randomly selected, by the at least one
computer processor, for assignment.
15. An article of manufacture for benchmarking pairing strategies
in a service center system comprising: a non-transitory processor
readable medium; and instructions stored on the medium; wherein the
instructions are configured to be readable from the medium by at
least one computer processor communicatively coupled to and
configured to operate in the service center system and thereby
cause the at least one computer processor to operate so as to:
receive a first plurality of results for a first plurality of cases
assigned to a plurality of agents using a first pairing strategy;
receive a second plurality of results for a second plurality of
cases assigned to the plurality of agents using a second pairing
strategy different from the first pairing strategy; determine a
difference in performance between the first and second pluralities
of results, wherein the difference in performance provides an
indication that assigning cases using the second pairing strategy
results in a performance gain for the service center system
attributable to the second pairing strategy, wherein the difference
in performance also provides an indication that optimizing
performance of the service center system is realized using the
second pairing strategy instead of the first pairing strategy; and
output the difference in performance between the first pairing
strategy and the second pairing strategy for benchmarking at least
the first pairing strategy and the second pairing strategy.
16. The article of manufacture of claim 15, wherein the first
pairing strategy is case assignment by a management authority.
17. The article of manufacture of claim 15, wherein the first
pairing strategy is a first-in, first out (FIFO) pairing
strategy.
18. The article of manufacture of claim 15, wherein the second
pairing strategy is a behavioral pairing strategy to optimize
performance of the service center system.
19. The article of manufacture of claim 15, wherein each case of
the second plurality of cases comprises a confidence level below a
threshold confidence level.
20. The article of manufacture of claim 15, wherein each case of
the first plurality of cases comprises a rationale from a
management authority and a confidence level above a threshold
confidence level.
Description
FIELD OF THE DISCLOSURE
This disclosure generally relates to customer service/contact
center case assignment, more particularly, to techniques for
collaborative and non-collaborative allocations of cases to agents
using behavioral pairing.
BACKGROUND OF THE DISCLOSURE
In some customer service centers, cases may be assigned to agents
(e.g., analysts, specialists) for servicing. For example, insurance
claims may be assigned to insurance adjusters or other agents for
subrogation or other processing; patients or other insureds may be
assigned to nurses, pharmacists, or other clinical support
specialists; debt collectors may be assigned to debtor cases; and
so on. These cases may be assigned in a variety of ways. In some
customer service centers (including, for example, workflow, case
management, or transaction processing service or support
organizations), cases may be assigned to agents based on time of
arrival. This strategy may be referred to as a "first-in,
first-out", "FIFO", or "round-robin" strategy. In some customer
service centers, management (e.g., managers or supervisors) may
assign cases to agents (including other types of specialists such
as those mentioned above), possibly with a particular rationale
based on information known to the management, such as information
about an agent's skills or historical performance. For some cases,
management may have low confidence in their assignments or lack
relevant information to make optimal assignments.
Also, in some customer contact centers, cases or contacts may be
assigned to agents for servicing. For example, a "lead list" of
contacts may be generated for each agent to contact (e.g., using an
outbound dialer). These contacts may be assigned to agents using a
FIFO strategy. In other environments, contacts may be assigned to
agents using other methods such as management-based
assignments.
In view of the foregoing, it may be understood that there may be
significant problems and shortcomings associated with current FIFO
or management-assigned strategies.
SUMMARY OF THE DISCLOSURE
Techniques for case allocation are disclosed. In one particular
embodiment, the techniques may be realized as a method for case
allocation comprising receiving, by at least one computer
processor, at least one case allocation allocated using a first
pairing strategy, and then reassigning, by the at least one
computer processor, the at least one case allocation using
behavioral pairing.
In accordance with other aspects of this particular embodiment, the
first pairing strategy is assigned by management.
In accordance with other aspects of this particular embodiment, the
first pairing strategy is a first-in, first-out (FIFO) pairing
strategy.
In accordance with other aspects of this particular embodiment, a
subsequent reassignment of the at least one case allocation using
the first pairing strategy may be received by the at least one
computer processor.
In accordance with other aspects of this particular embodiment, a
subsequent reversion of the at least one case allocation using the
first pairing strategy may be received by the at least one computer
processor.
In accordance with other aspects of this particular embodiment, a
plurality of case allocations allocated using the first pairing
strategy is received by the at least one computer processor, the
plurality of case allocations may be split by the at least one
computer processor into at least a first portion of cases and a
second portion of cases, and the second portion of case allocations
may be reassigned by the at least one computer processor using
behavioral pairing without reassigning the first portion of case
allocations.
In accordance with other aspects of this particular embodiment, a
difference in performance between the first portion of case
allocations and the second portion of case allocations may be
determined by the at least one computer processor.
In accordance with other aspects of this particular embodiment,
splitting the plurality of cases is based in part on at least one
rationale from management for at least one of the plurality of case
allocations.
In accordance with other aspects of this particular embodiment,
splitting the plurality of cases is based in part on at least one
confidence level from management for at least one of the plurality
of case allocations.
In another particular embodiment, the techniques may be realized as
a system for case allocation comprising at least one computer
processor configured to receive at least one case allocation
allocated using a first pairing strategy, and then reassign the at
least one case allocation using behavioral pairing. The system may
also comprise at least one memory, coupled to the at least one
computer processor, configured to provide the at least one computer
processor with instructions.
In another particular embodiment, the techniques may be realized as
an article of manufacture for case allocation comprising at least
one non-transitory computer processor readable medium and
instructions stored on the at least one medium, wherein the
instructions are configured to be readable from the at least one
medium by at least one computer processor and thereby cause the at
least one computer processor to operate so as to receive at least
one case allocation allocated using a first pairing strategy and
then reassign the at least one case allocation using behavioral
pairing.
The present disclosure will now be described in more detail with
reference to particular embodiments thereof as shown in the
accompanying drawings. While the present disclosure is described
below with reference to particular embodiments, it should be
understood that the present disclosure is not limited thereto.
Those of ordinary skill in the art having access to the teachings
herein will recognize additional implementations, modifications,
and embodiments, as well as other fields of use, which are within
the scope of the present disclosure as described herein, and with
respect to which the present disclosure may be of significant
utility.
BRIEF DESCRIPTION OF THE DRAWINGS
In order to facilitate a fuller understanding of the present
disclosure, reference is now made to the accompanying drawings, in
which like elements are referenced with like numerals. These
drawings should not be construed as limiting the present
disclosure, but are intended to be illustrative only.
FIG. 1 shows a flow diagram of a collaborative allocation system
according to embodiments of the present disclosure.
FIG. 2 shows a flow diagram of a collaborative allocation method
according to embodiments of the present disclosure.
FIG. 3 shows a schematic representation of case splits according to
embodiments of the present disclosure.
FIG. 4 shows a flow diagram of a non-collaborative allocation
system according to embodiments of the present disclosure.
FIG. 5 shows a flow diagram of a non-collaborative allocation
method according to embodiments of the present disclosure.
DETAILED DESCRIPTION
In some customer service centers, cases may be assigned to agents
(e.g., analysts, specialists) for servicing. For example, insurance
claims may be assigned to insurance adjusters or other agents for
subrogation or other processing; patients or other insureds may be
assigned to nurses, pharmacists, or other clinical support
specialists; debt collectors may be assigned to debtor cases; and
so on. These cases may be assigned in a variety of ways. In some
customer service centers (including, for example, workflow, case
management, or transaction processing service or support
organizations), cases may be assigned to agents based on time of
arrival. This strategy may be referred to as a "first-in,
first-out", "FIFO", or "round-robin" strategy. In some customer
service centers, management (e.g., managers or supervisors) may
assign cases to agents (including other types of specialists such
as those mentioned above), possibly with a particular rationale
based on information known to the management, such as information
about an agent's skills or historical performance. For some cases,
management may have low confidence in their assignments or lack
relevant information to make optimal assignments.
Also, in some customer contact centers, cases or contacts may be
assigned to agents for servicing. For example, a "lead list" of
contacts may be generated for each agent to contact (e.g., using an
outbound dialer). These contacts may be assigned to agents using a
FIFO strategy. In other environments, contacts may be assigned to
agents using other methods such as management-based
assignments.
In some embodiments, management assignments may be collaboratively
enhanced using an automated case assignment system, such as a
behavioral pairing module as described in U.S. patent Ser. No.
14/871,658, filed Sep. 30, 2015, now U.S. Pat. No. 9,300,802,
issued Mar. 29, 2016, and incorporated by reference herein. In this
way, a collaborative allocation system may leverage a big data,
artificial intelligence pairing solution (e.g., the behavioral
pairing module) with management expertise (e.g., a management
assignment module) to optimize case assignment, resulting in
increased performance in a customer service center. For example,
collaborative allocation or other uses of behavioral pairing of
cases may result in increased subrogation recoveries for insurance
claims, improved care for medical patients, improved debt
collection, and so on. In other embodiments, behavioral pairing and
management-based pairing may be performed separately in a
non-collaborative fashion.
In some embodiments, behavioral pairing may be performed "offline"
(e.g., not in real time) to assign cases, generate lead lists, or
perform other types of contact assignments using collaborative or
non-collaborative techniques.
Additionally, the improved performance of collaboratively-allocated
cases or non-collaboratively allocated cases as compared to
management-allocated cases may be precisely measurable as a gain
(e.g., 1%, 3%, 5%, etc.). In some embodiments, gain may be
precisely measured using a benchmarking module as described in U.S.
patent application Ser. No. 15/131,915, filed Apr. 18, 2016.
FIG. 1 depicts the workflow of a collaborative allocation system
100 according to some embodiments of the present disclosure.
Cases for assignment 110 may be received by, e.g., a management
assignment module 120 at a customer service center. The management
assignment module 120 may be provided solely by the customer
service center (including other types of customer service centers
and aforementioned support organizations), or it may be provided in
whole or in part as a component of a collaborative allocation
system.
The management assignment module 120 may output initial assignment
data 130. Initial assignment data 130 may include pairings of cases
with agents (including other types of agents and aforementioned
specialists), and it may include management rationale for these
pairings. For example, each pairing may have an associated score
representing management's confidence (e.g., certainty) in a
particular pairing. In some embodiments, each pairing may have one
or more associated reason codes or other codes indicating
management's reasons for a particular pairing (e.g., a good fit
with agent's skills or personality given information about the
agent known to management). Pairings may include an expected level
of time or effort (e.g., intensity) required to resolve the case.
Pairings may also take into account balancing caseload across
agents including agents' capacities to take on additional cases
with varying requirements for time or effort.
The initial assignment data 130 may be analyzed by a behavioral
pairing module 140 (or similar pairing engine). At this point, some
cases will be excluded (i.e., reserved or frozen) by management.
For example, if management has expressed high confidence or a
particular reason code for a case, or if the behavioral pairing
module 140 has determined low ability improve the initial
assignment, the behavioral pairing module 140 will not consider
this case for reassignment.
The remaining cases may be split into cases that may be reassigned
(e.g., an optimized or "on" group) and cases that may not be
reassigned (e.g., a control or "off" group). This split may be done
according to any of many possible splitting strategies. For
example, management may provide a seed to a pseudorandom number
generator, which may be used to randomly distributed cases into one
group or the other. In some embodiments, cases will be divided
evenly between the groups. In other embodiments, an uneven
distribution of cases may be used. For example, 80% of the cases
available for reassignment may be split into the optimized group,
while 20% of the cases available for reassignment may be split into
the control group. The technique used for splitting cases between
the groups may be designed to ensure transparency and fairness when
benchmarking performance.
Following the splitting of the cases, the cases in the optimized
group may be reassigned by the behavioral pairing module 140 or
similar automatic pairing techniques. In some embodiments, the
behavioral pairing module 140 may incorporate data about the agents
and the management (e.g., agent survey data 150A, management survey
data 150B, historical data 150C). The survey may include
self-assessment questions (e.g., which types of cases are you most
skilled at? Which types of cases do you prefer to handle? Which
stage of a case are you most skilled at? Which stage of a case do
you prefer to handle?). For management, survey questions may be
directed at understanding a manager's rationale for assigning
particular types of cases or cases at particular stages to
particular agents. Historical data may include information such as
historical case assignments and outcomes, case "scores" or other
case assessments prior to assignment, and other baseline
performance measurements. The behavioral pairing module 140 may
also search/analyze/process other data sources for information that
may be relevant to optimizing assignments and creating artificial
intelligence models. The behavioral pairing module 140 may account
for any stage of the case management process to optimize case
assignments, such as workflow, case management, transaction
processing, etc.
The behavioral pairing module 140 may output reassignment data 160,
which may include pairings from the optimized group that have been
reassigned to different agents. In some embodiments, the
reassignment data 160 may be reviewed by the management assignment
module 120, and the management assignment module 120 may optionally
output revised reassignment data 170. For example, the revised
reassignment data 170 may optionally "undo", revert, or otherwise
change some of the reassigned pairings based, for example, on
information known to management.
Subsequently, the benchmarking module 180 may measure the gain in
performance attributable to the collaboration between management
and the behavioral pairing module 140. The benchmarking module 180
may process the outcomes of each pairing to determine the relative
performance of cases in the optimized or "on" group, which were
collaborative allocated, against the performance of cases in the
control or "off" group, which were allocated solely by management.
The benchmarking module 180 may output performance measurements 190
(e.g., gain) or other information regarding the performance of the
collaborative allocation system 100.
The collaborative allocation system 100 may repeat this process as
new cases for assignment (e.g., cases for assignment 110) arrive or
otherwise become ready to be allocated among the agents. In some
embodiments, the management assignment module 120 or the behavioral
pairing module may process results from earlier iterations to
improve the management process (e.g., train managers regarding
certain rationales that were more or less effective than others) or
the behavioral pairing process (e.g., train or update the
artificial intelligence algorithms or models).
In some embodiments, the collaborative allocation system 100 may
operate "online" (e.g., in real time) as cases arrive at a queue or
as management assignments are made. In other embodiments, the
collaborative allocation system 100 may operate "offline" (e.g.,
not in real time), so that a group of cases may be reassigned or
otherwise allocated together.
FIG. 2 shows a flowchart of a collaborative allocation method 200
according to embodiments of the present disclosure. At block 210,
collaborative allocation method 200 may begin.
At block 210, preparatory information for collaborative allocation
may be processed. For example, an assignment or pairing module
(e.g., behavioral pairing module 140) may receive agent survey
data, management survey data, historical data, or other information
for processing in preparation for reassigning or otherwise
allocating cases to agents. Collaborative allocation method 200 may
proceed to block 220.
At block 220, initial assignment data (e.g., initial management
assignment data) may be received. In some embodiments, rationales
for management assignments may also be received. Collaborative
allocation method 200 may proceed to block 230.
At block 230, a portion of cases may be split out for reassignment,
while another portion of cases may be excluded (reserved, frozen,
or otherwise held back) from potential reassignment. In some
embodiments, these cases may also be excluded from benchmarking
measurements. Collaborative allocation method 200 may proceed to
block 240.
At block 240, the portion of cases split out for reassignment may
be reassigned. In some embodiments, reassignment may be performed
by a pairing module such as behavioral pairing module 140. In some
embodiments, reassignment data may be output or otherwise returned
for management review or further assignment. In some embodiments, a
portion of the cases split out for reassignment may be designated
to a control group and will not be reassigned. Collaborative
allocation method 200 may proceed to block 250.
At block 250, revisions to reassignments, if any, may be received.
In some embodiments, management may revise, revert, or otherwise
change the reassignments that were carried out by the pairing
module at block 240. Revised or reverted cases may be included or
excluded from benchmarking measurements. Collaborative allocation
method 200 may proceed to block 260.
At block 260, the relative performance of collaboratively-assigned
cases and management-assigned cases may be benchmarked or otherwise
measured. In some embodiments, results from the comparison may be
used to improve the pairing module (e.g., artificial intelligence
models of behavioral pairing module 140) or the rationales of
management for subsequent management assignments, or both.
Following block 260, collaborative allocation method 200 may end.
In some embodiments, collaborative allocation method 200 may return
to block 210 to begin allocating additional cases.
FIG. 3 depicts a schematic representation of case splits according
to embodiments of the present disclosure. As shown in FIG. 3, seven
agents may be assigned up to nine cases. Some cases may be
designated as "Ongoing" (e.g., cases that were previously assigned
but not yet complete). "Excluded" (i.e., frozen or held back) cases
are cases assigned to an agent that were determined to not be made
available for reassignment. "Management" cases are cases assigned
to an agent that were made available for reassignment but were
allocated to the control group. "Joint" cases are cases allocated
to the optimized group, which were jointly/collaboratively
reassigned and/or revised by management.
In the example of FIG. 3, seven agents (labeled 1 to 7 in the
"Agent" column) have a docket or queue of nine cases (labeled "Case
1" to "Case 9" in the header row). Agent 1's first case ("Case 1")
is identified by an "O" for Ongoing, and cases 2-9 have been split
for assignment or collaborative allocation: Cases 3 and 8 ("E")
have been excluded from collaborative allocation, and may
optionally be excluded from any benchmarking or relative
performance analysis. Cases 4, 5, and 7 ("M") have been assigned by
management, and may be benchmarked as being part of the control or
off cycle. Cases 2, 6, and 9 ("J") have been allocated by an
automated pairing strategy such as behavioral pairing, and may be
benchmarked as being part of the optimized or on cycle. In the case
of collaborative allocation, the optimized pairings may be made
jointly with management. In other embodiments, such as
non-collaborative allocation, the optimized pairings may be made
independently by the pairing strategy such as behavioral pairing,
without revision or reassignment by management. The remaining
agents Agent 2 to Agent 7 have been assigned or reassigned up to
nine available cases in a similar manner. As agents close cases in
their dockets or queues, and as more cases become available for
assignment, these new cases may be split for assignment or
reassignment among the available agents according to the
collaborative or non-collaborative allocation techniques in use for
this set of agents.
The outcome of each case may be associated with whether a case was
ongoing, excluded, management-assigned, or jointly-assigned using a
pairing strategy such as behavioral pairing. The relative
performance of different assignment methodologies may be
benchmarked or otherwise measured. For example, the performance
gain attributable to jointly-assigned cases using behavioral
pairing over management-assigned cases may be benchmarked.
FIG. 4 depicts the workflow of a non-collaborative allocation
system 400 according to some embodiments of the present
disclosure.
Cases for assignment 110 may be received at a contact center. The
cases may be split into two or more groups for assignment by
different strategies. In some embodiments, a portion of cases may
be assigned randomly, on a FIFO basis, by management, or other case
allocation techniques. A second portion of cases may be assigned
using a pairing strategy such as behavioral pairing. In some
embodiments, as in the example of FIG. 4, a first portion of cases
may be received by management assignment module 120, and a second
portion of cases may be received by behavioral pairing module
140.
The management assignment module 120 may output management
assignment data 410. Management assignment data 410 may include
pairings of cases with agents, and it may include management
rationale for these pairings. For example, each pairing may have an
associated score representing management's confidence (e.g.,
certainty) in a particular pairing. In some embodiments, each
pairing may have one or more associated reason codes or other codes
indicating management's reasons for a particular pairing (e.g., a
good fit with agent's skills or personality given information about
the agent known to management). Pairings may include an expected
level of time or effort (e.g., intensity) required to resolve the
case. Pairings may also take into account balancing caseload across
agents including agents' capacities to take on additional cases
with varying requirements for time or effort.
The behavioral pairing module 140 may output behavioral pairing
assignment data 420. In some embodiments, the behavioral pairing
module 140 may incorporate data about the agents and the management
(e.g., agent survey data 150A, management survey data 150B,
historical data 150C). The survey may include self-assessment
questions (e.g., which types of cases are you most skilled at?
Which types of cases do you prefer to handle? Which stage of a case
are you most skilled at? Which stage of a case do you prefer to
handle?). For management, survey questions may be directed at
understanding a manager's rationale for assigning particular types
of cases or cases at particular stages to particular agents.
Historical data may include information such as historical case
assignments and outcomes, case "scores" or other case assessments
prior to assignment, and other baseline performance measurements.
The behavioral pairing module 140 may also search/analyze/process
other data sources for information that may be relevant to
optimizing assignments and creating artificial intelligence
models.
Subsequently, the benchmarking module 180 may measure the gain in
performance attributable to the behavioral pairing module 140 as
compared to the management assignment module (or other assignment
process such as a random or FIFO process). The benchmarking module
180 may process the outcomes of each pairing to determine the
relative performance of cases in the optimized group, which were
allocated solely using behavioral pairing, against the performance
of cases in the control group, which were allocated solely by
management. The benchmarking module 180 may output performance
measurements 190 or other information regarding the performance of
the non-collaborative allocation system 400.
The non-collaborative allocation system 400 may repeat this process
as new cases for assignment (e.g., cases for assignment 110) arrive
or otherwise become ready to be allocated among the agents. In some
embodiments, the management assignment module 120 or the behavioral
pairing module 140 may process results from earlier iterations to
improve the management process (e.g., train managers regarding
certain rationales that were more or less effective than others) or
the behavioral pairing process (e.g., train or update the
artificial intelligence algorithms or models).
In some embodiments, the non-collaborative allocation system 400
may operate "online" (e.g., in real time) as cases arrive at a
queue or as management assignments are made. In other embodiments,
the non-collaborative allocation system 400 may operate "offline"
(e.g., not in real time), so that a group of cases may be
reassigned or otherwise allocated together.
FIG. 5 shows a flow diagram of a non-collaborative allocation
method according to embodiments of the present disclosure. At block
510, non-collaborative allocation method 500 may begin.
At block 510, preparatory information for non-collaborative
allocation may be processed. For example, an assignment or pairing
module (e.g., behavioral pairing module 140) may receive agent
survey data, management survey data, historical data, or other
information for processing in preparation for assigning or
otherwise allocating cases to agents. Non-collaborative allocation
method 500 may proceed to block 520.
At block 520, cases may be split into first and second portions of
one or more cases. Non-collaborative allocation method 500 may
proceed to block 530.
At block 530, assignment data may be received for the portion of
cases split out for management assignment (or, e.g., random or FIFO
assignment). Non-collaborative allocation method 500 may proceed to
block 540.
At block 540, the second portion of cases may be assigned using a
pairing strategy such as behavioral pairing (BP). Non-collaborative
allocation method 500 may proceed to block 550.
At block 550, the relative performance of BP-assigned cases and
management-assigned cases may be benchmarked or otherwise measured.
In some embodiments, results from the comparison may be used to
improve the pairing module (e.g., artificial intelligence models of
behavioral pairing module 140) or the rationales of management for
subsequent management assignments, or both.
Following block 550, non-collaborative allocation method 500 may
end. In some embodiments, non-collaborative allocation method 500
may return to block 510 to begin allocating additional cases.
At this point it should be noted that collaborative and
non-collaborative allocation using behavioral pairing in accordance
with the present disclosure as described above may involve the
processing of input data and the generation of output data to some
extent. This input data processing and output data generation may
be implemented in hardware or software. For example, specific
electronic components may be employed in a collaborative and
non-collaborative allocation module, behavioral pairing module,
benchmarking module, and/or similar or related circuitry for
implementing the functions associated with collaborative and
non-collaborative allocation using behavioral pairing, such as in a
workflow management system, contact center system, case management
system, etc. in accordance with the present disclosure as described
above. Alternatively, one or more processors operating in
accordance with instructions may implement the functions associated
with collaborative and non-collaborative allocation using
behavioral pairing in accordance with the present disclosure as
described above. If such is the case, it is within the scope of the
present disclosure that such instructions may be stored on one or
more non-transitory computer processor readable storage media
(e.g., a magnetic disk or other storage medium), or transmitted to
one or more computer processors via one or more signals embodied in
one or more carrier waves.
The present disclosure is not to be limited in scope by the
specific embodiments described herein. Indeed, other various
embodiments of and modifications to the present disclosure, in
addition to those described herein, will be apparent to those of
ordinary skill in the art from the foregoing description and
accompanying drawings. Thus, such other embodiments and
modifications are intended to fall within the scope of the present
disclosure. Further, although the present disclosure has been
described herein in the context of at least one particular
implementation in at least one particular environment for at least
one particular purpose, those of ordinary skill in the art will
recognize that its usefulness is not limited thereto and that the
present disclosure may be beneficially implemented in any number of
environments for any number of purposes. Accordingly, the claims
set forth below should be construed in view of the full breadth and
spirit of the present disclosure as described herein.
* * * * *
References